On 06-Nov-97 Steve Lamont wrote:
>To subscribers of the xforms list from spl@szechuan.ucsd.edu (Steve Lamont) :
>
>> I just have no idea what could cause that behaviour. I see no
>> relation between that pressed key and the pointer. "I feel so
>> stupid!"
>
>Do you have some other event handler registered?
No other handle! Only the "idle_callback"!
>What I meant was that you've probably either got
>a bad array index somewhere that's going beyond the end of an array
>and causing the program to write where it isn't intended to write or
>you are perhaps referencing a bad pointer -- perhaps you've done a
>free() somewhere but are still inadvertently using the now invalidated
>pointer.
No arrays in the program and I think I checked all pointers but I'll do it
again!
>> >Where does the core dump occur? Can you do a traceback with a
>> >debugger (the gdb `where' command will do this).
The traceback again delivers fl_get_choice_text as last called function.
it look something like that (output of the gdb frontend ddd!)
#7 0x8048efb in ___crt_dummy__ ()
#6 0x80492d0 in main (argc=1, argv=0xbffff668) at ToGo.c:136
#5 0x40029c64 in fl_do_forms()
#4 0x400295f0 in do_interaction_step()
#2 0x40029269 in fl_handle_automatic()
#1 0x804917d in redraw (event=0x4005f9b8, data=0x0) at ToGo.c:98
#0 0x4001dd2e in fl_get_choice_text()
But that's not yet of big use for me, i'm not smart enough for that.
>> Sorry I get no core dumped. Can I force the program to do so? Or what does
>it
>> mean when core is dumped or not? (fatal error <-> bad error)
>
>There's no `core' file produced? What kind of system are you on?
>Some versions of Unix have a `limit' command in the shell.
It's a Linux system with Kernel 2.0.25 running on an amd 486.
No limit command!
>Alternatively, run your program from within your debugger. Refer to
>your system's manual pages for instructions. With a good debugger you
>can set "watch points" to cause the program to stop at the exact point
>a variable is modified. You should, of course, compile the program
>with all optimization turned off and with debugging symbols turned
>on. Again, refer to your manual pages for details.
OK, i'll try to do so and report some results as soon as I get some ;-)
Thank you
best regards
Bjoern
_________________________________________________
To unsubscribe, send the message "unsubscribe" to
xforms-request@bob.usuf2.usuhs.mil or see
http://bob.usuf2.usuhs.mil/mailserv/xforms.html
Xforms Home Page: http://bragg.phys.uwm.edu/xforms
List Archive: http://bob.usuf2.usuhs.mil/mailserv/list-archives/